The proposed study addresses critical gaps in research that have impeded efforts to improve intervention and address mental health disparities for diverse, urban, low-SES infants and their families. Compelling longitudinal data have shown that insecure infant attachment is linked to later psychopathology (e.g., Sroufe et al., 2005); thus, it is crucially important to understand precursors of attachment that can be targeted in preventive intervention. Meta-analysis, however, shows that this goal has proven elusive, particularly in low-SES families (De Wolff & van IJzendoorn, 1997). There is a great public health need for effective intervention for infants in low-SES families because low-SES infants tend to face multiple risk factors (McLoyd, 1998) and insecure attachment is thought to be a risk factor for psychopathology in the context of other risk factors (DeKlyen & Greenberg, 2008). The proposed study has three interrelated aims: First, the proposed research challenges current paradigms for studying precursors of attachment by examining a novel conceptualization of parental care giving: Secure Base Provision (SBP; Cassidy, Woodhouse, et al., 2005; Woodhouse & Cassidy, 2009) - the degree to which a parent is able to ultimately meet an infant's attachment needs, even when demonstrating high levels of insensitive parenting behavior. The construct of SBP is rooted in Bowlby's (1988) emphasis on the attachment figure's role in providing a secure base for exploration to which the child can return when needed. SBP avoids emphasizing the importance of certain culturally bound parenting practices found in white, middle class samples (e.g., sweet tone of voice, moment-to-moment affective attunement), and does not pathologize other parenting practices (e.g., no-nonsense parenting). The proposed study will test the relative predictive utility of SBP and maternal sensitivity-the degree to which a mother is able to accurately interpret infant needs and respond appropriately and promptly (Ainsworth, et al., 1971; NICHD, 1999)-in predicting later attachment. Second, the proposed research will break new ground by using a longitudinal design with a racially/ethnically diverse, urban, low-SES sample to examine the relative predictive utility of SBP vs. sensitivity in predicting infant stress reactivity (as indexed by cortisol and alpha-amylase), physiological indicators of emotion regulation (i.e., RSA), and behavioral problems (e.g., internalizing and externalizing problems), so as to better understand links between care giving and later emotion regulation. Third, in order to better understand the correlates of maternal care giving behavior, we will examine the role of maternal stress reactivity and maternal physiological regulation in care giving. The notion that parental emotion regulation plays an important role in responsive parenting (Dix, 1991) has received surprisingly little empirical attention (Lorber & O'Leary, 2005), yet could be useful in understanding how to best intervene with mothers. In sum, successful completion of the three aims will inform the development of interventions for diverse, low-SES families by extending current theory and research on parental care giving, infant emotion regulation, and infant behavioral problems.